Jacknife statistics of model performance and relative importance of each variable.

For each species, the table shows the area under the curve (AUC) and regularized training gain (Gain). For each variable first value correspond to the gain of a model fitted using all variables except the focal one. The more important variable according to this criterion is marked with *. The second value corresponds to the gain of a model fitted using just the focal variable. The more important variable according to this criterion is marked with †, (see methods for details). Abbreviations as in table 1.